platform/core/uifw/dali-adaptor.git
10 years agoRemoved legacy resource tracking / logging 61/17761/1
Richard Underhill [Mon, 10 Mar 2014 10:48:26 +0000 (10:48 +0000)]
Removed legacy resource tracking / logging

[Issue#] N/A

[Problem]

[Cause]

[Solution]

Change-Id: If08b6c30617f7fc62dce1ba76436eaa54b5c79ee
Signed-off-by: Richard Underhill <r.underhill@partner.samsung.com>
10 years ago(VirtualKeyboard) Adding Guard around ImfManager::Get() so if fails functions are... 04/17704/1
Agnelo Vaz [Thu, 6 Mar 2014 16:26:04 +0000 (16:26 +0000)]
(VirtualKeyboard) Adding Guard around ImfManager::Get() so if fails functions are not called.

[Issue#] N/A
[Problem] N/A
[Cause] N/A
[Solution]
[Verification] Build Repo

Change-Id: Ieb0fe71064b95ba14ee6363d57051b5a4b87f4f2
Signed-off-by: Paul Wisbey <p.wisbey@samsung.com>
10 years agoRemove null statement 03/17703/1
Francisco Santos [Thu, 6 Mar 2014 15:17:14 +0000 (15:17 +0000)]
Remove null statement

[Issue#] (N/A)

[Problem]

[Cause]

[Solution]

Signed-off-by: Paul Wisbey <p.wisbey@samsung.com>
Change-Id: Ibfe3ee92ebe5a537b8b155d7d0acf7da89c23cee

10 years agoRemove redundant Core::SendEvent() & NotificationEvent 27/17227/1
Paul Wisbey [Tue, 4 Mar 2014 11:29:46 +0000 (11:29 +0000)]
Remove redundant Core::SendEvent() & NotificationEvent

[Issue#]   N/A
[Problem]  Notification event is redundant
[Cause]    Previously there was no way of telling Core to process events,
except by sending an event.
[Solution] Use the ProcessEvents() method, which was separated from Core::SendEvent()

Signed-off-by: Paul Wisbey <p.wisbey@samsung.com>
Change-Id: Ice003fb098d337fe84ec40581a19e5da5c94fcc3

10 years agoUse separated Core::QueueEvent & ProcessEvents methods (instead of SendEvent)
Paul Wisbey [Thu, 6 Feb 2014 18:39:59 +0000 (18:39 +0000)]
Use separated Core::QueueEvent & ProcessEvents methods (instead of SendEvent)

[Issue#]   N/A
[Problem]  Adaptor is calling SendEvent multiple times for each raw touch event,
causing extra effort with message queuing, size negotiation etc.
[Cause]    SendEvent was doing two different jobs i.e. queueing & processing.
[Solution] Use the separated QueueEvent & ProcessEvents methods

10 years ago(KeyEvents) Removing unused cursorOffset numberOfChars from KeyEvent
Agnelo Vaz [Thu, 27 Feb 2014 19:54:20 +0000 (19:54 +0000)]
(KeyEvents) Removing unused cursorOffset numberOfChars from KeyEvent

[Issue#] N/A
[Problem] CAPI Freeze so cleaning up first
[Cause] N/A
[Solution]
[Verification] Build Repo

Change-Id: Ia82332ea9b86b998068b0617641f4a0e2441f115

10 years agoAbility to log performance makers to ftrace for Tizen SPI tool
Nick Holland [Fri, 21 Feb 2014 14:13:44 +0000 (14:13 +0000)]
Ability to log performance makers to ftrace for Tizen SPI tool

[Issue] N/A

[Problem]
When using ftrace to analyse what the system is doing there is no way to see
what task Dali is doing. E.g. when does it start running its update-thread
in relation to the V-SYNC and event processing.

[Cause] No code

[Solution]

Added extra bitmask option to DALI_LOG_PERFORMANCE environment variable,
Example:

DALI_LOG_PERFORMANCE=4 dali-demo

cat /sys/kernel/debug/tracing/trace

 TASK-PID    CPU#    TIMESTAMP  FUNCTION

  <...>-28189 [001] 269275.222762: tracing_mark_write: SPI_EV_DALI_PROCESS_EVENT_START
  <...>-28189 [001] 269275.222812: tracing_mark_write: SPI_EV_DALI_PROCESS_EVENT_END
  <...>-28196 [000] 269275.239498: tracing_mark_write: SPI_EV_DALI_V_SYNC
  <...>-28194 [005] 269275.239527: tracing_mark_write: SPI_EV_DALI_UPDATE_START
  <...>-28194 [005] 269275.240401: tracing_mark_write: SPI_EV_DALI_UPDATE_END
  <...>-28196 [000] 269275.256192: tracing_mark_write: SPI_EV_DALI_V_SYNC
  <...>-28194 [005] 269275.256228: tracing_mark_write: SPI_EV_DALI_UPDATE_START
  <...>-28194 [005] 269275.256330: tracing_mark_write: SPI_EV_DALI_UPDATE_END

The SPI_EV_ prefix is added to allow SPI tool to detect the markers.

Change-Id: Idec9714308aa19a9f1ef509057f6ba7d227c4649

10 years ago(ImageLoading) Added new method to get closest image size
David Steele [Wed, 19 Feb 2014 20:26:57 +0000 (20:26 +0000)]
(ImageLoading) Added new method to get closest image size

[Issue#]       N/A
[Problem]      No synchronous method to get closest image size.
Jpeg loader will give different size if requested, as it implements image scaling.
Want to have image actor size set to the natural size from creation, rather than
only after loading.
[Cause]        Not implemented
[Solution]
Changed signature of BitmapLoader header function to pass in ImageAttributes.
Changed meaning of header function - if a non-zero size is set, then the header
function should work out what size will be returned for the requested size and
orientation correction parameters (as would be done in the actual loader function)

refactored jpeg loader to write new header method to handle scaling and orientation
correction

Changed png header to account for cropping

[Verification] Build Repo

Change-Id: I1c9b814aafacff70f089df0735ab5bb6a108cfcc
Signed-off-by: David Steele <david.steele@partner.samsung.com>
10 years ago[SRUK] Initial copy from Tizen 2.2 version
Paul Wisbey [Mon, 3 Mar 2014 18:55:55 +0000 (18:55 +0000)]
[SRUK] Initial copy from Tizen 2.2 version

Signed-off-by: Paul Wisbey <p.wisbey@samsung.com>
Change-Id: I666580b4d50f1476f4117f93f27901e61f94bcf8

10 years agoInitial empty repository
Sehong Na [Sun, 16 Feb 2014 23:48:41 +0000 (15:48 -0800)]
Initial empty repository